In [17]:
atri = network(0)
import os
## Set directory
os.chdir('/hpc/group/pbenfeylab/CheWei/CW_data/genesys')
import networkx as nx
from genesys_evaluate import *
%matplotlib inline
%config InlineBackend.figure_format = 'retina'
import warnings
# Suppress all warning messages
warnings.filterwarnings("ignore", category=DeprecationWarning)
## Conda Env pytorch-gpu on DCC
print(torch.__version__)
print(sc.__version__)
1.13.0.post200 1.9.1
## Genes considered/used (shared among samples)
gene_list = pd.read_csv('./gene_list_1108.csv')
with open("./genesys_root_data.pkl", 'rb') as file_handle:
data = pickle.load(file_handle)
batch_size = 2000
dataset = Root_Dataset(data['X_test'], data['y_test'])
loader = DataLoader(dataset,
batch_size = batch_size,
shuffle = True, drop_last=True)
input_size = data['X_train'].shape[1]
## 10 cell types
output_size = 10
embedding_dim = 256
hidden_dim = 256
n_layers = 2
device = "cpu"
path = "./"
model = ClassifierLSTM(input_size, output_size, embedding_dim, hidden_dim, n_layers).to(device)
model.load_state_dict(torch.load(path+"best_ALL_1130_continue.pth", map_location=torch.device('cpu')))
model = model
model.eval()
ClassifierLSTM(
(fc1): Sequential(
(0): Linear(in_features=17513, out_features=256, bias=True)
(1): Dropout(p=0.2, inplace=False)
(2): GaussianNoise()
)
(fc): Sequential(
(0): ReLU()
(1): Linear(in_features=512, out_features=512, bias=True)
(2): ReLU()
(3): Linear(in_features=512, out_features=10, bias=True)
)
(lstm): LSTM(256, 256, num_layers=2, batch_first=True, dropout=0.2, bidirectional=True)
(dropout): Dropout(p=0.2, inplace=False)
(b_to_z): DBlock(
(fc1): Linear(in_features=512, out_features=256, bias=True)
(fc2): Linear(in_features=512, out_features=256, bias=True)
(fc_mu): Linear(in_features=256, out_features=512, bias=True)
(fc_logsigma): Linear(in_features=256, out_features=512, bias=True)
)
(bz2_infer_z1): DBlock(
(fc1): Linear(in_features=1024, out_features=256, bias=True)
(fc2): Linear(in_features=1024, out_features=256, bias=True)
(fc_mu): Linear(in_features=256, out_features=512, bias=True)
(fc_logsigma): Linear(in_features=256, out_features=512, bias=True)
)
(z1_to_z2): DBlock(
(fc1): Linear(in_features=512, out_features=256, bias=True)
(fc2): Linear(in_features=512, out_features=256, bias=True)
(fc_mu): Linear(in_features=256, out_features=512, bias=True)
(fc_logsigma): Linear(in_features=256, out_features=512, bias=True)
)
(z_to_x): Decoder(
(fc1): Linear(in_features=512, out_features=256, bias=True)
(fc2): Linear(in_features=256, out_features=256, bias=True)
(fc3): Linear(in_features=256, out_features=17513, bias=True)
)
)
classes = ['Columella', 'Lateral Root Cap', 'Phloem', 'Xylem', 'Procambium', 'Pericycle', 'Endodermis', 'Cortex', 'Atrichoblast', 'Trichoblast']
class2num = {c: i for (i, c) in enumerate(classes)}
num2class = {i: c for (i, c) in enumerate(classes)}
cts = ['Atrichoblast','Trichoblast','Cortex','Endodermis','Pericycle','Procambium','Xylem','Phloem','Lateral Root Cap','Columella']
ctw = np.zeros((len(cts), 17513, 17513))
## number of cells sampled from the atlas
batch_size = 2000
## GRN for the transition t7 to t9
for ct in cts:
print(ct)
cws = np.zeros((len(loader), 17513, 17513))
with torch.no_grad():
for i, sample in enumerate(loader):
x = sample['x'].to(device)
y = sample['y'].to(device)
y_label = [num2class[i] for i in y.tolist()]
pred_h = model.init_hidden(batch_size)
tfrom = model.generate_next(x, pred_h, 7).to('cpu').detach().numpy()
cfrom = tfrom[np.where(np.array(y_label)==ct)[0],:]
pred_h = model.init_hidden(batch_size)
tto = model.generate_next(x, pred_h, 9).to('cpu').detach().numpy()
cto = tto[np.where(np.array(y_label)==ct)[0],:]
cw = torch.linalg.lstsq(torch.tensor(cfrom), torch.tensor(cto)).solution.detach().numpy()
cws[i] = cw
## Calculate mean across number of repeats
cwm = np.mean(cws, axis=0)
ctw[cts.index(ct)] = cwm
Atrichoblast Trichoblast Cortex Endodermis Pericycle Procambium Xylem Phloem Lateral Root Cap Columella
# Save the array to disk
np.save('genesys_ctw_t7-t9.npy', ctw)
ctw = np.load('genesys_ctw_t7-t9.npy')
## Calculate z-scores
ctw_z = np.zeros((len(cts), 17513, 17513))
for i in range(len(cts)):
ctw_z[i] = (ctw[i] - np.mean(ctw[i])) / np.std(ctw[i])
## Filtering based on z-scores (with no weights)
ctw_f = np.zeros((len(cts), 17513, 17513))
## z-score threshold (keep values > mean + threshold*std)
threshold=3
for i in range(len(cts)):
ctw_f[i] = np.abs(ctw_z[i]) > threshold
wanted_TFs = pd.read_csv("./Kay_TF_thalemine_annotations.csv")
## Make TF names unique and assign preferred names
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT2G33880"]="WOX9"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT2G45160"]="SCL27"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G04410"]="NAC78"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT3G29035"]="ORS1"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT2G02540"]="ZHD3"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT3G16500"]="IAA26"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G09740"]="HAG5"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT4G24660"]="ZHD2"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G46880"]="HDG5"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G28420"]="RLT1"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G14580"]="BLJ"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT3G45260"]="BIB"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT2G02070"]="RVN"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT2G28160"]="FIT"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G68360"]="GIS3"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G20640"]="NLP4"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G05550"]="VFP5"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT3G59470"]="FRF1"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G15150"]="HAT7"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G14750"]="WER"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G75710"]="BRON"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G74500"]="TMO7"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT2G12646"]="RITF1"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT3G48100"]="ARR5"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT4G16141"]="GATA17L"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G65640"]="NFL"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G62700"]="VND5"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT4G36160"]="VND2"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G66300"]="VND3"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT1G12260"]="VND4"
wanted_TFs['Name'][wanted_TFs['GeneID']=="AT5G62380"]="VND6"
pd.Series(wanted_TFs['Name']).value_counts().head(5)
NAC001 1 PRE5 1 MYB118 1 MYB21 1 MYB0 1 Name: Name, dtype: int64
TFidx = []
for i in wanted_TFs['GeneID']:
if i in gene_list['features'].tolist():
TFidx.append(np.where(gene_list['features']==i)[0][0])
TFidx = np.sort(np.array(TFidx))
def network(i):
## No weights
adj_nw = ctw_f[i]
## Weighted
adj = ctw[i]*ctw_f[i]
## TF only
adj = adj[np.ix_(TFidx,TFidx)]
adj_nw = adj_nw[np.ix_(TFidx,TFidx)]
## Remove no connect
regidx = np.sort(np.array(pd.Series(np.where(adj_nw==True)[0]).value_counts().index[pd.Series(np.where(adj_nw==True)[0]).value_counts()>=1]))
taridx = np.sort(np.array(pd.Series(np.where(adj_nw==True)[1]).value_counts().index[pd.Series(np.where(adj_nw==True)[1]).value_counts()>=1]))
## Reciprocol
keepidx = np.sort(np.array(list(set(regidx).intersection(taridx))))
#keepidx = np.sort(np.array(list(set(regidx).union(taridx))))
TFID = np.array(gene_list['features'][TFidx])[keepidx].tolist()
## TF name to keep
TFname = []
for i in np.array(gene_list['features'][TFidx])[keepidx]:
TFname.append(wanted_TFs['Name'][np.where(wanted_TFs['GeneID']==i)[0][0]])
adj = adj[np.ix_(keepidx,keepidx)]
# Create a NetworkX graph for non-directed edges
G = nx.Graph() # supports directed edges and allows for multiple edges between the same pair of nodes
# Add nodes to the graph
num_nodes = adj.shape[0]
for i, name in enumerate(TFname):
G.add_node(i, name=name)
# Add edges to the graph with weights
for i in range(num_nodes):
for j in range(num_nodes):
weight = adj[i, j]
if weight != 0:
G.add_edge(i, j, weight=abs(weight), distance=1/abs(weight))
## Measures the extent to which how close a node is to all other nodes in the network, considering the shortest paths or geodesic distances between nodes
closeness_centrality = nx.closeness_centrality(G, distance='distance')
## Measures the extent to which a node that are not only well-connected but also connected to other well-connected nodes.
eigenvector_centrality = nx.eigenvector_centrality(G)
# Create a NetworkX graph for diected edges
G = nx.MultiDiGraph() # supports directed edges and allows for multiple edges between the same pair of nodes
# Add nodes to the graph
num_nodes = adj.shape[0]
for i, name in enumerate(TFname):
G.add_node(i, name=name)
# Add edges to the graph with weights
for i in range(num_nodes):
for j in range(num_nodes):
weight = adj[i, j]
if weight != 0:
G.add_edge(i, j, weight=weight)
## Measures the number of connections (edges) each node has
degree_centrality = nx.degree_centrality(G)
# Calculate outgoing centrality
out_centrality = nx.out_degree_centrality(G)
# Calculate incoming centrality
in_centrality = nx.in_degree_centrality(G)
## Measures the extent to which a node lies on the shortest paths between other nodes.
betweenness_centrality = nx.betweenness_centrality(G, weight='weight')
## Non_Reciprocal Out centrality
# Visualize the graph
pos = nx.spring_layout(G) # Positions of the nodes
# Node colors based on weighted betweenness centrality
node_colors = [out_centrality[node] for node in G.nodes()]
# Node sizes based on weighted betweenness centrality
node_sizes = [out_centrality[node] * 1000 for node in G.nodes()]
# Get the edge weights as a dictionary
edge_weights = nx.get_edge_attributes(G, 'weight')
edge_colors = ['red' if weight > 0 else 'blue' for (_, _, weight) in G.edges(data='weight')]
# Scale the edge weights to desired linewidths
max_weight = max(edge_weights.values())
edge_widths = [float(edge_weights[edge]) / max_weight for edge in G.edges]
# Draw the graph
nx.draw(G, pos=pos, node_color=node_colors, node_size=node_sizes, with_labels=False, width=edge_widths, edge_color=edge_colors)
# Add node labels
labels = {node: G.nodes[node]['name'] for node in G.nodes}
nx.draw_networkx_labels(G, pos=pos, labels=labels, font_size=8)
# Add a colorbar to show the weighted betweenness centrality color mapping
sm = plt.cm.ScalarMappable(cmap='viridis', norm=plt.Normalize(vmin=min(node_colors), vmax=max(node_colors)))
sm.set_array([])
plt.colorbar(sm)
# Show the plot
plt.show()
dc = pd.DataFrame.from_dict(degree_centrality, orient='index', columns=['degree_centrality'])
oc = pd.DataFrame.from_dict(out_centrality, orient='index', columns=['out_centrality'])
ic = pd.DataFrame.from_dict(in_centrality, orient='index', columns=['in_centrality'])
bc = pd.DataFrame.from_dict(betweenness_centrality, orient='index', columns=['betweenness_centrality'])
cc = pd.DataFrame.from_dict(closeness_centrality, orient='index', columns=['closeness_centrality'])
ec = pd.DataFrame.from_dict(eigenvector_centrality, orient='index', columns=['eigenvector_centrality'])
df = pd.concat([dc,oc,ic,bc,cc,ec], axis=1)
df.index =TFname
df = df.sort_values('betweenness_centrality', ascending=False)
return(df)
atri = network(0)
tri = network(1)
cor = network(2)
end = network(3)
per = network(4)
pro = network(5)
xyl = network(6)
phl = network(7)
lrc = network(8)
col = network(9)
atri.columns = ['atri_degree_centrality','atri_out_centrality','atri_in_centrality','atri_betweenness_centrality','atri_closeness_centrality','atri_eigenvector_centrality']
tri.columns = ['tri_degree_centrality','tri_out_centrality','tri_in_centrality','tri_betweenness_centrality','tri_closeness_centrality','tri_eigenvector_centrality']
cor.columns = ['cor_degree_centrality','cor_out_centrality','cor_in_centrality','cor_betweenness_centrality','cor_closeness_centrality','cor_eigenvector_centrality']
end.columns = ['end_degree_centrality','end_out_centrality','end_in_centrality','end_betweenness_centrality','end_closeness_centrality','end_eigenvector_centrality']
per.columns = ['per_degree_centrality','per_out_centrality','per_in_centrality','per_betweenness_centrality','per_closeness_centrality','per_eigenvector_centrality']
pro.columns = ['pro_degree_centrality','pro_out_centrality','pro_in_centrality','pro_betweenness_centrality','pro_closeness_centrality','pro_eigenvector_centrality']
xyl.columns = ['xyl_degree_centrality','xyl_out_centrality','xyl_in_centrality','xyl_betweenness_centrality','xyl_closeness_centrality','xyl_eigenvector_centrality']
phl.columns = ['phl_degree_centrality','phl_out_centrality','phl_in_centrality','phl_betweenness_centrality','phl_closeness_centrality','phl_eigenvector_centrality']
lrc.columns = ['lrc_degree_centrality','lrc_out_centrality','lrc_in_centrality','lrc_betweenness_centrality','lrc_closeness_centrality','lrc_eigenvector_centrality']
col.columns = ['col_degree_centrality','col_out_centrality','col_in_centrality','col_betweenness_centrality','col_closeness_centrality','col_eigenvector_centrality']
## Indentify main regulators in each net work
tff = []
tff = tff + atri[atri['atri_betweenness_centrality']>0].index.tolist()
tff = tff + tri[tri['tri_betweenness_centrality']>0].index.tolist()
tff = tff + lrc[lrc['lrc_betweenness_centrality']>0].index.tolist()
tff = tff + cor[cor['cor_betweenness_centrality']>0].index.tolist()
tff = tff + end[end['end_betweenness_centrality']>0].index.tolist()
tff = tff + per[per['per_betweenness_centrality']>0].index.tolist()
tff = tff + pro[pro['pro_betweenness_centrality']>0].index.tolist()
tff = tff + xyl[xyl['xyl_betweenness_centrality']>0].index.tolist()
tff = tff + phl[phl['phl_betweenness_centrality']>0].index.tolist()
tff = tff + col[col['col_betweenness_centrality']>0].index.tolist()
tf_occurance = pd.DataFrame(pd.Series(tff).value_counts(), columns=['tf_occurance'])
tf_spec = pd.concat([tf_occurance, atri, tri, lrc, cor, end, per, pro, xyl, phl, col], axis=1)
tf_spec = tf_spec.fillna(0)
## Epidermis (atri, tri, lrc)
celltype1='atri'
celltype2='tri'
celltype3='lrc'
ts = tf_spec[tf_spec['tf_occurance']==3][[celltype1+'_betweenness_centrality', celltype2+'_betweenness_centrality', celltype3+'_betweenness_centrality', celltype1+'_out_centrality', celltype2+'_out_centrality', celltype3+'_out_centrality', celltype1+'_in_centrality', celltype2+'_in_centrality', celltype3+'_in_centrality']]
tso = (ts > 0)
ts['centrality_count'] = tso.sum(axis=1)
ts['centrality_sum'] = ts.sum(axis=1)
ts[ts['centrality_count']==9].sort_values(['centrality_count','centrality_sum'], ascending=False)
| atri_betweenness_centrality | tri_betweenness_centrality | lrc_betweenness_centrality | atri_out_centrality | tri_out_centrality | lrc_out_centrality | atri_in_centrality | tri_in_centrality | lrc_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| AT2G28710 | 0.915490 | 0.018175 | 0.002137 | 0.424855 | 0.216 | 0.004274 | 0.456647 | 0.074667 | 0.220085 | 9 | 11.332330 |
| KAN | 0.677172 | 0.000043 | 0.000014 | 0.017341 | 0.056 | 0.051282 | 0.225434 | 0.098667 | 0.044872 | 9 | 10.170823 |
| AT5G22890 | 0.013513 | 0.000349 | 0.121452 | 0.144509 | 0.024 | 0.008547 | 0.156069 | 0.066667 | 0.023504 | 9 | 9.558610 |
## atri, tri
celltype1='atri'
celltype2='tri'
ts = tf_spec[tf_spec['tf_occurance']==2][[celltype1+'_betweenness_centrality', celltype2+'_betweenness_centrality', celltype1+'_out_centrality', celltype2+'_out_centrality', celltype1+'_in_centrality', celltype2+'_in_centrality']]
tso = (ts > 0)
ts['centrality_count'] = tso.sum(axis=1)
ts['centrality_sum'] = ts.sum(axis=1)
ts[ts['centrality_count']==6].sort_values(['centrality_count','centrality_sum'], ascending=False)
| atri_betweenness_centrality | tri_betweenness_centrality | atri_out_centrality | tri_out_centrality | atri_in_centrality | tri_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|---|---|---|
| LRL3 | 0.879551 | 0.871879 | 0.736994 | 0.768000 | 0.156069 | 0.930667 | 6 | 10.343160 |
| RHD6 | 0.675480 | 0.926474 | 0.320809 | 0.330667 | 0.075145 | 0.669333 | 6 | 8.997908 |
| WRKY61 | 0.912909 | 0.756606 | 0.193642 | 0.141333 | 0.265896 | 0.578667 | 6 | 8.849053 |
| NAC6 | 0.913043 | 0.000093 | 0.358382 | 0.216000 | 0.511561 | 0.045333 | 6 | 8.044412 |
| HRS1 | 0.472481 | 0.001747 | 0.196532 | 0.061333 | 0.182081 | 0.264000 | 6 | 7.178173 |
| HFR1 | 0.417550 | 0.021234 | 0.193642 | 0.042667 | 0.115607 | 0.186667 | 6 | 6.977366 |
| MYB50 | 0.026012 | 0.000813 | 0.130058 | 0.018667 | 0.352601 | 0.106667 | 6 | 6.634817 |
| WRKY4 | 0.005772 | 0.027622 | 0.034682 | 0.109333 | 0.060694 | 0.029333 | 6 | 6.267436 |
## Atrichoblast specific
celltype = 'atri'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| atri_betweenness_centrality | atri_out_centrality | atri_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| AT4G01350 | 0.00578 | 0.083815 | 0.075145 | 3 | 3.164740 |
| HB24 | 0.00578 | 0.176301 | 0.213873 | 3 | 3.395954 |
| PUX2 | 0.00289 | 0.005780 | 0.031792 | 3 | 3.040462 |
## Trichoblast specific
celltype = 'tri'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| tri_betweenness_centrality | tri_out_centrality | tri_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| RSL2 | 0.913319 | 0.056000 | 0.634667 | 3 | 4.603986 |
| AT4G09100 | 0.817291 | 0.760000 | 0.424000 | 3 | 5.001291 |
| IAA31 | 0.771258 | 0.218667 | 0.194667 | 3 | 4.184592 |
| AT5G56200 | 0.369561 | 0.138667 | 0.194667 | 3 | 3.702895 |
| KNAT3 | 0.332720 | 0.085333 | 0.064000 | 3 | 3.482053 |
| AT4G39160 | 0.109476 | 0.189333 | 0.064000 | 3 | 3.362809 |
| AT3G53370 | 0.090781 | 0.256000 | 0.114667 | 3 | 3.461447 |
| RAP2.11 | 0.070595 | 0.085333 | 0.208000 | 3 | 3.363929 |
| RSL4 | 0.046403 | 0.136000 | 0.264000 | 3 | 3.446403 |
| HB23 | 0.017583 | 0.069333 | 0.117333 | 3 | 3.204250 |
| AT2G05160 | 0.017098 | 0.085333 | 0.136000 | 3 | 3.238431 |
| AT2G20030 | 0.013077 | 0.045333 | 0.152000 | 3 | 3.210410 |
| FBH3 | 0.010467 | 0.061333 | 0.101333 | 3 | 3.173134 |
| AT2G37120 | 0.004520 | 0.072000 | 0.056000 | 3 | 3.132520 |
| AT2G14760 | 0.002610 | 0.013333 | 0.013333 | 3 | 3.029276 |
| MC2 | 0.002610 | 0.061333 | 0.090667 | 3 | 3.154610 |
| bZIP23 | 0.002602 | 0.037333 | 0.064000 | 3 | 3.103936 |
| NPR1 | 0.001569 | 0.013333 | 0.090667 | 3 | 3.105569 |
| AT3G05760 | 0.000991 | 0.026667 | 0.077333 | 3 | 3.104991 |
| AT2G27580 | 0.000684 | 0.056000 | 0.058667 | 3 | 3.115351 |
| WRKY42 | 0.000043 | 0.029333 | 0.149333 | 3 | 3.178709 |
## LRC specific
celltype = 'lrc'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| lrc_betweenness_centrality | lrc_out_centrality | lrc_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| TTG2 | 0.889603 | 0.085470 | 0.094017 | 3 | 4.069090 |
| MYB51 | 0.777604 | 0.021368 | 0.025641 | 3 | 3.824612 |
| IAA1 | 0.558658 | 0.074786 | 0.072650 | 3 | 3.706094 |
| CRF3 | 0.539669 | 0.155983 | 0.117521 | 3 | 3.813174 |
| PLT1 | 0.358622 | 0.380342 | 0.019231 | 3 | 3.758195 |
| SPT | 0.358361 | 0.104701 | 0.025641 | 3 | 3.488703 |
| ARF16 | 0.339272 | 0.072650 | 0.087607 | 3 | 3.499529 |
| AT1G09250 | 0.308639 | 0.027778 | 0.038462 | 3 | 3.374879 |
| GATA2 | 0.191658 | 0.243590 | 0.019231 | 3 | 3.454478 |
| ATCTH | 0.125556 | 0.213675 | 0.057692 | 3 | 3.396923 |
| AT1G26680 | 0.120033 | 0.452991 | 0.014957 | 3 | 3.587982 |
| TRP1 | 0.113701 | 0.051282 | 0.032051 | 3 | 3.197034 |
| OFP6 | 0.110018 | 0.158120 | 0.034188 | 3 | 3.302325 |
| ASL1 | 0.082267 | 0.143162 | 0.089744 | 3 | 3.315173 |
| HAT7 | 0.011484 | 0.185897 | 0.032051 | 3 | 3.229433 |
| GRF2 | 0.008167 | 0.074786 | 0.023504 | 3 | 3.106458 |
| MYB23 | 0.004274 | 0.055556 | 0.138889 | 3 | 3.198718 |
| AT2G42300 | 0.002137 | 0.036325 | 0.042735 | 3 | 3.081197 |
| AT2G29065 | 0.001290 | 0.059829 | 0.029915 | 3 | 3.091034 |
| NAC052 | 0.000943 | 0.070513 | 0.108974 | 3 | 3.180430 |
| RR12 | 0.000764 | 0.087607 | 0.023504 | 3 | 3.111875 |
| E2F1 | 0.000645 | 0.038462 | 0.027778 | 3 | 3.066884 |
| HAM3 | 0.000384 | 0.100427 | 0.025641 | 3 | 3.126453 |
| AT5G62610 | 0.000325 | 0.085470 | 0.008547 | 3 | 3.094342 |
| WRKY65 | 0.000229 | 0.087607 | 0.012821 | 3 | 3.100656 |
| AT1G69030 | 0.000037 | 0.081197 | 0.025641 | 3 | 3.106874 |
## Columella specific
celltype = 'col'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| col_betweenness_centrality | col_out_centrality | col_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| ARF10 | 0.468511 | 0.042994 | 0.176752 | 3 | 3.688256 |
| SCL1 | 0.466436 | 0.098726 | 0.184713 | 3 | 3.749876 |
| RAT5 | 0.462749 | 0.014331 | 0.076433 | 3 | 3.553513 |
| SNI1 | 0.007952 | 0.073248 | 0.098726 | 3 | 3.179926 |
| AT1G76580 | 0.006367 | 0.039809 | 0.286624 | 3 | 3.332800 |
| AT1G21780 | 0.005775 | 0.019108 | 0.076433 | 3 | 3.101317 |
| HSF A4A | 0.003332 | 0.027070 | 0.144904 | 3 | 3.175307 |
| AT2G44430 | 0.001732 | 0.098726 | 0.054140 | 3 | 3.154598 |
| BZIP60 | 0.001615 | 0.074841 | 0.046178 | 3 | 3.122634 |
| AT3G08505 | 0.001592 | 0.009554 | 0.046178 | 3 | 3.057325 |
| TTG1 | 0.001592 | 0.074841 | 0.014331 | 3 | 3.090764 |
| EIL3 | 0.001592 | 0.025478 | 0.038217 | 3 | 3.065287 |
| WRKY35 | 0.001587 | 0.052548 | 0.036624 | 3 | 3.090759 |
| RHA1 | 0.001582 | 0.007962 | 0.001592 | 3 | 3.011136 |
| BBX28 | 0.000742 | 0.015924 | 0.022293 | 3 | 3.038958 |
| CHR11 | 0.000510 | 0.031847 | 0.101911 | 3 | 3.134268 |
| RR1 | 0.000373 | 0.200637 | 0.073248 | 3 | 3.274259 |
| AT3G52250 | 0.000183 | 0.065287 | 0.055732 | 3 | 3.121202 |
| ATRX | 0.000140 | 0.128981 | 0.085987 | 3 | 3.215108 |
| SNL2 | 0.000137 | 0.090764 | 0.038217 | 3 | 3.129118 |
| SPL1 | 0.000114 | 0.066879 | 0.019108 | 3 | 3.086102 |
| IDD1 | 0.000112 | 0.062102 | 0.036624 | 3 | 3.098838 |
| AT5G12400 | 0.000066 | 0.052548 | 0.070064 | 3 | 3.122677 |
| TLP7 | 0.000061 | 0.090764 | 0.030255 | 3 | 3.121080 |
| AT1G74250 | 0.000056 | 0.031847 | 0.074841 | 3 | 3.106744 |
| CDC5 | 0.000033 | 0.039809 | 0.025478 | 3 | 3.065320 |
| TLP10 | 0.000025 | 0.068471 | 0.046178 | 3 | 3.114675 |
| HSL1 | 0.000020 | 0.028662 | 0.117834 | 3 | 3.146517 |
| AT1G50410 | 0.000013 | 0.030255 | 0.038217 | 3 | 3.068484 |
| EIL1 | 0.000008 | 0.130573 | 0.012739 | 3 | 3.143320 |
| BRM | 0.000005 | 0.074841 | 0.025478 | 3 | 3.100324 |
| HMG | 0.000005 | 0.090764 | 0.027070 | 3 | 3.117839 |
## Ground tissue
celltype1='cor'
celltype2='end'
ts = tf_spec[tf_spec['tf_occurance']==2][[celltype1+'_betweenness_centrality', celltype2+'_betweenness_centrality', celltype1+'_out_centrality', celltype2+'_out_centrality', celltype1+'_in_centrality', celltype2+'_in_centrality']]
tso = (ts > 0)
ts['centrality_count'] = tso.sum(axis=1)
ts['centrality_sum'] = ts.sum(axis=1)
ts[ts['centrality_count']==6].sort_values(['centrality_count','centrality_sum'], ascending=False)
| cor_betweenness_centrality | end_betweenness_centrality | cor_out_centrality | end_out_centrality | cor_in_centrality | end_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|---|---|---|
| AT1G05710 | 0.483670 | 0.023826 | 0.808642 | 0.116022 | 0.962963 | 0.296961 | 6 | 8.692084 |
| MYB122 | 0.002523 | 0.004275 | 0.246914 | 0.016575 | 0.163580 | 0.157459 | 6 | 6.591325 |
| IDD4 | 0.003736 | 0.000023 | 0.200617 | 0.027624 | 0.243827 | 0.031768 | 6 | 6.507596 |
| RAX2 | 0.000143 | 0.016544 | 0.237654 | 0.088398 | 0.043210 | 0.051105 | 6 | 6.437054 |
| ZFN1 | 0.000143 | 0.000827 | 0.089506 | 0.037293 | 0.101852 | 0.004144 | 6 | 6.233765 |
| GATA27 | 0.000382 | 0.000258 | 0.077160 | 0.015193 | 0.040123 | 0.045580 | 6 | 6.178698 |
| AT2G47890 | 0.000057 | 0.071598 | 0.046296 | 0.023481 | 0.030864 | 0.004144 | 6 | 6.176440 |
| BBX21 | 0.000029 | 0.000059 | 0.033951 | 0.017956 | 0.003086 | 0.026243 | 6 | 6.081324 |
## Cortex specific
celltype = 'cor'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| cor_betweenness_centrality | cor_out_centrality | cor_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| LAF1 | 0.095574 | 0.262346 | 0.842593 | 3 | 4.200512 |
| JKD | 0.029068 | 0.466049 | 0.108025 | 3 | 3.603142 |
| JAZ6 | 0.012871 | 0.212963 | 0.354938 | 3 | 3.580772 |
| BZS1 | 0.008562 | 0.330247 | 0.259259 | 3 | 3.598068 |
| HB5 | 0.007797 | 0.438272 | 0.132716 | 3 | 3.578785 |
| GLK2 | 0.006221 | 0.012346 | 0.209877 | 3 | 3.228443 |
| AT1G72210 | 0.004357 | 0.172840 | 0.256173 | 3 | 3.433370 |
| KNAT5 | 0.003077 | 0.030864 | 0.098765 | 3 | 3.132706 |
| NSI | 0.001682 | 0.018519 | 0.027778 | 3 | 3.047978 |
| NAC084 | 0.001424 | 0.055556 | 0.006173 | 3 | 3.063152 |
| AT2G42660 | 0.000965 | 0.021605 | 0.148148 | 3 | 3.170718 |
| COL4 | 0.000707 | 0.132716 | 0.154321 | 3 | 3.287744 |
| ABF2 | 0.000487 | 0.070988 | 0.018519 | 3 | 3.089994 |
| AT3G61180 | 0.000468 | 0.027778 | 0.046296 | 3 | 3.074542 |
| ABS2 | 0.000382 | 0.033951 | 0.003086 | 3 | 3.037419 |
| JMJ18 | 0.000315 | 0.009259 | 0.009259 | 3 | 3.018834 |
| ASIL2 | 0.000296 | 0.009259 | 0.006173 | 3 | 3.015728 |
| AT4G16150 | 0.000296 | 0.030864 | 0.052469 | 3 | 3.083630 |
| GATA26 | 0.000248 | 0.009259 | 0.083333 | 3 | 3.092841 |
| RAP2.7 | 0.000239 | 0.117284 | 0.030864 | 3 | 3.148387 |
| IAA16 | 0.000229 | 0.067901 | 0.018519 | 3 | 3.086649 |
| WRKY69 | 0.000182 | 0.117284 | 0.080247 | 3 | 3.197712 |
| LRP1 | 0.000172 | 0.058642 | 0.040123 | 3 | 3.098937 |
| IDD7 | 0.000105 | 0.009259 | 0.185185 | 3 | 3.194550 |
| PIL5 | 0.000048 | 0.027778 | 0.012346 | 3 | 3.040171 |
| AT2G38300 | 0.000048 | 0.030864 | 0.089506 | 3 | 3.120418 |
| AT5G24320 | 0.000048 | 0.067901 | 0.027778 | 3 | 3.095727 |
| MYB14 | 0.000029 | 0.021605 | 0.021605 | 3 | 3.043239 |
| AT1G21580 | 0.000029 | 0.012346 | 0.061728 | 3 | 3.074103 |
| PAT1 | 0.000029 | 0.049383 | 0.070988 | 3 | 3.120399 |
| SIGA | 0.000029 | 0.018519 | 0.003086 | 3 | 3.021634 |
| FRS8 | 0.000019 | 0.012346 | 0.018519 | 3 | 3.030883 |
| AGL17 | 0.000019 | 0.009259 | 0.015432 | 3 | 3.024710 |
| LBD39 | 0.000019 | 0.012346 | 0.015432 | 3 | 3.027797 |
| LZF1 | 0.000019 | 0.030864 | 0.012346 | 3 | 3.043229 |
| AP2 | 0.000019 | 0.037037 | 0.033951 | 3 | 3.071007 |
| SIGE | 0.000019 | 0.021605 | 0.067901 | 3 | 3.089525 |
| ZFN3 | 0.000019 | 0.033951 | 0.024691 | 3 | 3.058661 |
| BZO2H1 | 0.000019 | 0.009259 | 0.012346 | 3 | 3.021624 |
| RBR1 | 0.000019 | 0.003086 | 0.006173 | 3 | 3.009278 |
| AT3G13040 | 0.000010 | 0.024691 | 0.037037 | 3 | 3.061738 |
| RGL3 | 0.000010 | 0.021605 | 0.027778 | 3 | 3.049392 |
| FRS2 | 0.000010 | 0.012346 | 0.021605 | 3 | 3.033960 |
| NPH4 | 0.000010 | 0.015432 | 0.015432 | 3 | 3.030874 |
| AT4G01280 | 0.000010 | 0.012346 | 0.027778 | 3 | 3.040133 |
| ERF15 | 0.000010 | 0.129630 | 0.046296 | 3 | 3.175935 |
| ETR1 | 0.000010 | 0.009259 | 0.043210 | 3 | 3.052479 |
| AGL42 | 0.000010 | 0.129630 | 0.043210 | 3 | 3.172849 |
| DBP1 | 0.000010 | 0.030864 | 0.012346 | 3 | 3.043219 |
| ALY2 | 0.000010 | 0.006173 | 0.015432 | 3 | 3.021614 |
## Endodermis specific
celltype = 'end'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| end_betweenness_centrality | end_out_centrality | end_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| bZIP58 | 0.467489 | 0.096685 | 0.310773 | 3 | 3.874947 |
| AT4G00940 | 0.379573 | 0.082873 | 0.194751 | 3 | 3.657197 |
| MYB68 | 0.267803 | 0.020718 | 0.174033 | 3 | 3.462554 |
| ERF73 | 0.016785 | 0.001381 | 0.029006 | 3 | 3.047171 |
| BLJ | 0.009432 | 0.116022 | 0.081492 | 3 | 3.206945 |
| AGL102 | 0.008068 | 0.013812 | 0.436464 | 3 | 3.458344 |
| TIFY8 | 0.004293 | 0.038674 | 0.143646 | 3 | 3.186613 |
| AT5G58620 | 0.001494 | 0.042818 | 0.125691 | 3 | 3.170002 |
| chr31 | 0.000556 | 0.234807 | 0.178177 | 3 | 3.413539 |
| AT1G61990 | 0.000252 | 0.002762 | 0.004144 | 3 | 3.007158 |
| TOC1 | 0.000145 | 0.004144 | 0.002762 | 3 | 3.007051 |
| BIB | 0.000092 | 0.004144 | 0.034530 | 3 | 3.038766 |
| COL3 | 0.000032 | 0.002762 | 0.001381 | 3 | 3.004176 |
| DEL2 | 0.000021 | 0.002762 | 0.013812 | 3 | 3.016596 |
| AT4G38340 | 0.000013 | 0.016575 | 0.290055 | 3 | 3.306643 |
| AMS | 0.000010 | 0.002762 | 0.051105 | 3 | 3.053877 |
| COP1 | 0.000004 | 0.016575 | 0.013812 | 3 | 3.030391 |
| SAP7 | 0.000002 | 0.024862 | 0.016575 | 3 | 3.041438 |
## Stele
celltype1='per'
celltype2='pro'
celltype3='xyl'
celltype4='phl'
ts = tf_spec[tf_spec['tf_occurance']==4][[celltype1+'_betweenness_centrality', celltype2+'_betweenness_centrality', celltype3+'_betweenness_centrality', celltype4+'_betweenness_centrality', celltype1+'_out_centrality', celltype2+'_out_centrality', celltype3+'_out_centrality', celltype4+'_out_centrality', celltype1+'_in_centrality', celltype2+'_in_centrality', celltype3+'_in_centrality', celltype4+'_in_centrality']]
tso = (ts > 0)
ts['centrality_count'] = tso.sum(axis=1)
ts['centrality_sum'] = ts.sum(axis=1)
ts[ts['centrality_count']==12].sort_values(['centrality_count','centrality_sum'], ascending=False)
| per_betweenness_centrality | pro_betweenness_centrality | xyl_betweenness_centrality | phl_betweenness_centrality | per_out_centrality | pro_out_centrality | xyl_out_centrality | phl_out_centrality | per_in_centrality | pro_in_centrality | xyl_in_centrality | phl_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| IAA11 | 0.007843 | 0.001321 | 0.041726 | 0.000117 | 0.064018 | 0.100592 | 0.309677 | 0.106132 | 0.116998 | 0.019724 | 0.741935 | 0.090802 | 12 | 13.600886 |
| RMA2 | 0.005347 | 0.000058 | 0.012359 | 0.000096 | 0.036424 | 0.086785 | 0.270968 | 0.104953 | 0.121413 | 0.003945 | 0.335484 | 0.063679 | 12 | 13.041510 |
## Pericycle
celltype = 'per'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| per_betweenness_centrality | per_out_centrality | per_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| MYBC1 | 0.991288 | 0.570640 | 0.272627 | 3 | 4.834555 |
| LBD14 | 0.959133 | 0.036424 | 0.014349 | 3 | 4.009906 |
| AT2G14880 | 0.640940 | 0.040839 | 0.054084 | 3 | 3.735863 |
| IDD11 | 0.280223 | 0.162252 | 0.048565 | 3 | 3.491039 |
| HDA3 | 0.083967 | 0.015453 | 0.016556 | 3 | 3.115976 |
| AT2G20100 | 0.044365 | 0.048565 | 0.054084 | 3 | 3.147014 |
| GATA23 | 0.020000 | 0.098234 | 0.022075 | 3 | 3.140310 |
| ULT1 | 0.006701 | 0.069536 | 0.006623 | 3 | 3.082860 |
| CDF3 | 0.003302 | 0.013245 | 0.093819 | 3 | 3.110366 |
| BHLH32 | 0.002582 | 0.108168 | 0.054084 | 3 | 3.164834 |
| AT1G75490 | 0.002200 | 0.024283 | 0.037528 | 3 | 3.064010 |
| GATA16 | 0.001400 | 0.090508 | 0.009934 | 3 | 3.101842 |
| PHV | 0.001104 | 0.008830 | 0.075055 | 3 | 3.084989 |
| BBX30 | 0.001104 | 0.001104 | 0.026490 | 3 | 3.028698 |
| AT2G40200 | 0.001099 | 0.024283 | 0.065121 | 3 | 3.090503 |
| ERF10 | 0.000583 | 0.030905 | 0.001104 | 3 | 3.032592 |
| SHR | 0.000009 | 0.101545 | 0.011038 | 3 | 3.112591 |
| HB20 | 0.000009 | 0.107064 | 0.007726 | 3 | 3.114799 |
| ZML2 | 0.000001 | 0.036424 | 0.015453 | 3 | 3.051878 |
## Procambium
celltype = 'pro'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| pro_betweenness_centrality | pro_out_centrality | pro_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| REV | 0.021267 | 0.063116 | 0.282051 | 3 | 3.366435 |
| AT4G25410 | 0.001949 | 0.045365 | 0.035503 | 3 | 3.082817 |
| AS1 | 0.000799 | 0.035503 | 0.013807 | 3 | 3.050109 |
| AT2G39020 | 0.000725 | 0.031558 | 0.007890 | 3 | 3.040173 |
| BES1 | 0.000686 | 0.031558 | 0.072978 | 3 | 3.105223 |
| GRP2 | 0.000674 | 0.019724 | 0.102564 | 3 | 3.122962 |
| CCA1 | 0.000651 | 0.041420 | 0.005917 | 3 | 3.047988 |
| AT1G28310 | 0.000635 | 0.017751 | 0.067061 | 3 | 3.085448 |
| AT3G03550 | 0.000468 | 0.023669 | 0.078895 | 3 | 3.103032 |
| AT2G38090 | 0.000273 | 0.009862 | 0.057199 | 3 | 3.067334 |
| TCP20 | 0.000261 | 0.031558 | 0.025641 | 3 | 3.057460 |
| CIP8 | 0.000144 | 0.013807 | 0.015779 | 3 | 3.029730 |
| GAI | 0.000144 | 0.049310 | 0.011834 | 3 | 3.061288 |
| HB18 | 0.000125 | 0.029586 | 0.088757 | 3 | 3.118468 |
| FRS12 | 0.000105 | 0.005917 | 0.019724 | 3 | 3.025746 |
| AGL80 | 0.000097 | 0.005917 | 0.007890 | 3 | 3.013904 |
| AT2G47210 | 0.000097 | 0.031558 | 0.021696 | 3 | 3.053352 |
| AT2G18090 | 0.000090 | 0.033531 | 0.009862 | 3 | 3.043482 |
| NF-YB10 | 0.000070 | 0.025641 | 0.023669 | 3 | 3.049380 |
| AT5G51790 | 0.000047 | 0.005917 | 0.023669 | 3 | 3.029633 |
| RGA1 | 0.000047 | 0.019724 | 0.017751 | 3 | 3.037522 |
| AGL18 | 0.000043 | 0.035503 | 0.001972 | 3 | 3.037518 |
| AL2 | 0.000043 | 0.019724 | 0.011834 | 3 | 3.031601 |
| AT4G22820 | 0.000039 | 0.023669 | 0.023669 | 3 | 3.047376 |
| TRFL6 | 0.000035 | 0.031558 | 0.005917 | 3 | 3.037510 |
| ARF9 | 0.000027 | 0.053254 | 0.025641 | 3 | 3.078923 |
| ATL5 | 0.000027 | 0.061144 | 0.001972 | 3 | 3.063144 |
| WRKY21 | 0.000019 | 0.059172 | 0.005917 | 3 | 3.065108 |
| GATA1 | 0.000019 | 0.009862 | 0.001972 | 3 | 3.011854 |
| AT5G09240 | 0.000019 | 0.013807 | 0.007890 | 3 | 3.021716 |
| ASL9 | 0.000016 | 0.017751 | 0.005917 | 3 | 3.023684 |
| AT4G31420 | 0.000016 | 0.043393 | 0.017751 | 3 | 3.061160 |
| FVE | 0.000016 | 0.023669 | 0.011834 | 3 | 3.035519 |
| AT4G25210 | 0.000016 | 0.043393 | 0.023669 | 3 | 3.067077 |
| POB1 | 0.000016 | 0.039448 | 0.019724 | 3 | 3.059187 |
| NF-YC4 | 0.000012 | 0.019724 | 0.013807 | 3 | 3.033542 |
| ASHH3 | 0.000012 | 0.015779 | 0.009862 | 3 | 3.025653 |
| RGL1 | 0.000012 | 0.001972 | 0.007890 | 3 | 3.009874 |
| ARIA | 0.000012 | 0.045365 | 0.019724 | 3 | 3.065100 |
| AGL26 | 0.000012 | 0.019724 | 0.003945 | 3 | 3.023680 |
| HMGB1 | 0.000008 | 0.047337 | 0.033531 | 3 | 3.080876 |
| ILR3 | 0.000008 | 0.037475 | 0.023669 | 3 | 3.061152 |
| HK2 | 0.000008 | 0.011834 | 0.003945 | 3 | 3.015787 |
| BPC6 | 0.000008 | 0.025641 | 0.005917 | 3 | 3.031566 |
| AT3G60080 | 0.000008 | 0.047337 | 0.019724 | 3 | 3.067069 |
| ASHH1 | 0.000008 | 0.035503 | 0.021696 | 3 | 3.057207 |
| MAF1 | 0.000004 | 0.033531 | 0.015779 | 3 | 3.049314 |
| NF-YC1 | 0.000004 | 0.001972 | 0.001972 | 3 | 3.003949 |
| AT5G09460 | 0.000004 | 0.023669 | 0.003945 | 3 | 3.027617 |
| STH | 0.000004 | 0.021696 | 0.003945 | 3 | 3.025645 |
| MYB109 | 0.000004 | 0.003945 | 0.001972 | 3 | 3.005921 |
| AGL21 | 0.000004 | 0.023669 | 0.001972 | 3 | 3.025645 |
| HTA10 | 0.000004 | 0.017751 | 0.001972 | 3 | 3.019728 |
| BPC4 | 0.000004 | 0.025641 | 0.011834 | 3 | 3.037479 |
| HSF3 | 0.000004 | 0.013807 | 0.003945 | 3 | 3.017755 |
| AIP3 | 0.000004 | 0.015779 | 0.007890 | 3 | 3.023673 |
| NF-YB1 | 0.000004 | 0.015779 | 0.001972 | 3 | 3.017755 |
## Xylem
celltype = 'xyl'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| xyl_betweenness_centrality | xyl_out_centrality | xyl_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| AT1G66810 | 0.967365 | 0.580645 | 0.780645 | 3 | 5.328655 |
| LBD31 | 0.946753 | 0.264516 | 0.703226 | 3 | 4.914495 |
| JLO | 0.943067 | 0.064516 | 0.845161 | 3 | 4.852744 |
| MYB83 | 0.876707 | 0.425806 | 0.793548 | 3 | 5.096062 |
| VND5 | 0.837160 | 0.090323 | 0.819355 | 3 | 4.746837 |
| VND4 | 0.830540 | 0.096774 | 0.838710 | 3 | 4.766024 |
| AT1G68810 | 0.820486 | 0.864516 | 0.619355 | 3 | 5.304357 |
| ZHD3 | 0.801131 | 0.896774 | 0.825806 | 3 | 5.523712 |
| AT3G10760 | 0.766820 | 0.245161 | 0.077419 | 3 | 4.089401 |
| MYB52 | 0.747047 | 0.251613 | 0.748387 | 3 | 4.747047 |
| IAA6 | 0.740302 | 0.083871 | 0.083871 | 3 | 3.908044 |
| MYB85 | 0.665061 | 0.316129 | 0.541935 | 3 | 4.523125 |
| VND1 | 0.496858 | 0.180645 | 0.077419 | 3 | 3.754922 |
| VND6 | 0.430163 | 0.458065 | 0.116129 | 3 | 4.004357 |
| GATA5 | 0.329493 | 0.019355 | 0.045161 | 3 | 3.394009 |
| AT2G22200 | 0.244198 | 0.025806 | 0.122581 | 3 | 3.392585 |
| MYB46 | 0.138961 | 0.974194 | 0.864516 | 3 | 4.977671 |
| AT5G46910 | 0.134018 | 0.212903 | 0.483871 | 3 | 3.830792 |
| BT1 | 0.132090 | 0.329032 | 0.025806 | 3 | 3.486929 |
| HB31 | 0.122832 | 0.877419 | 0.038710 | 3 | 4.038961 |
| SPL2 | 0.116129 | 0.038710 | 0.012903 | 3 | 3.167742 |
| AT2G28200 | 0.100461 | 0.316129 | 0.438710 | 3 | 3.855300 |
| AT3G10470 | 0.065396 | 0.141935 | 0.445161 | 3 | 3.652493 |
| VND3 | 0.032551 | 0.219355 | 0.554839 | 3 | 3.806745 |
| GIF3 | 0.032132 | 0.116129 | 0.219355 | 3 | 3.367616 |
| PIF4 | 0.030792 | 0.077419 | 0.225806 | 3 | 3.334018 |
| AT3G17100 | 0.012652 | 0.090323 | 0.070968 | 3 | 3.173942 |
| OFP10 | 0.012568 | 0.038710 | 0.154839 | 3 | 3.206116 |
| SAC51 | 0.006452 | 0.006452 | 0.148387 | 3 | 3.161290 |
| AT1G24040 | 0.006452 | 0.090323 | 0.148387 | 3 | 3.245161 |
| VND2 | 0.005614 | 0.406452 | 0.787097 | 3 | 4.199162 |
| AT1G68200 | 0.000922 | 0.206452 | 0.109677 | 3 | 3.317051 |
## Phloem
celltype = 'phl'
cs = tf_spec[tf_spec['tf_occurance']==1][[celltype+'_betweenness_centrality', celltype+'_out_centrality',celltype+'_in_centrality']]
cso = (cs > 0)
cs['centrality_count'] = cso.sum(axis=1)
cs['centrality_sum'] = cs.sum(axis=1)
cs[cs['centrality_count']==3].sort_values(['centrality_count',celltype+'_betweenness_centrality'], ascending=False)
| phl_betweenness_centrality | phl_out_centrality | phl_in_centrality | centrality_count | centrality_sum | |
|---|---|---|---|---|---|
| AT3G06220 | 0.054877 | 0.008255 | 0.022406 | 3 | 3.085538 |
| NAC057 | 0.023728 | 0.379717 | 0.379717 | 3 | 3.783162 |
| AT5G63700 | 0.017423 | 0.121462 | 0.201651 | 3 | 3.340536 |
| AT5G41380 | 0.013562 | 0.301887 | 0.487028 | 3 | 3.802477 |
| WOX2 | 0.013359 | 0.011792 | 0.001179 | 3 | 3.026330 |
| ... | ... | ... | ... | ... | ... |
| NAC78 | 0.000001 | 0.021226 | 0.023585 | 3 | 3.044813 |
| AT1G58220 | 0.000001 | 0.014151 | 0.011792 | 3 | 3.025945 |
| NFXL2 | 0.000001 | 0.010613 | 0.008255 | 3 | 3.018869 |
| AT1G01920 | 0.000001 | 0.007075 | 0.007075 | 3 | 3.014152 |
| KAN2 | 0.000001 | 0.031840 | 0.024764 | 3 | 3.056605 |
92 rows × 5 columns
gene = 'SHR'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
| tf_occurance | end_degree_centrality | end_out_centrality | end_in_centrality | end_closeness_centrality | end_eigenvector_centrality | per_degree_centrality | per_out_centrality | per_in_centrality | per_betweenness_centrality | per_closeness_centrality | per_eigenvector_centrality | pro_degree_centrality | pro_out_centrality | pro_in_centrality | pro_closeness_centrality | pro_eigenvector_centrality | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SHR | 1.0 | 0.002762 | 0.001381 | 0.001381 | 0.000172 | 0.002296 | 0.112583 | 0.101545 | 0.011038 | 0.000009 | 0.000376 | 0.051305 | 0.057199 | 0.035503 | 0.021696 | 0.00019 | 0.029499 |
gene = 'BLJ'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
| tf_occurance | end_degree_centrality | end_out_centrality | end_in_centrality | end_betweenness_centrality | end_closeness_centrality | end_eigenvector_centrality | |
|---|---|---|---|---|---|---|---|
| BLJ | 1.0 | 0.197514 | 0.116022 | 0.081492 | 0.009432 | 0.000315 | 0.105867 |
gene = 'JKD'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
| tf_occurance | cor_degree_centrality | cor_out_centrality | cor_in_centrality | cor_betweenness_centrality | cor_closeness_centrality | cor_eigenvector_centrality | end_degree_centrality | end_out_centrality | end_in_centrality | end_closeness_centrality | end_eigenvector_centrality | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| JKD | 1.0 | 0.574074 | 0.466049 | 0.108025 | 0.029068 | 0.000409 | 0.135451 | 0.150552 | 0.09116 | 0.059392 | 0.000301 | 0.089831 |
gene = 'RVN'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
| end_degree_centrality | end_out_centrality | end_in_centrality | end_closeness_centrality | end_eigenvector_centrality | |
|---|---|---|---|---|---|
| RVN | 0.125691 | 0.030387 | 0.095304 | 0.00026 | 0.085214 |
gene = 'BIB'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
| tf_occurance | end_degree_centrality | end_out_centrality | end_in_centrality | end_betweenness_centrality | end_closeness_centrality | end_eigenvector_centrality | |
|---|---|---|---|---|---|---|---|
| BIB | 1.0 | 0.038674 | 0.004144 | 0.03453 | 0.000092 | 0.000193 | 0.039742 |
gene = 'IME'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
gene = 'MYB66'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
gene = 'GL2'
tf_spec[tf_spec.index==gene][tf_spec[tf_spec.index==gene].columns[tf_spec[tf_spec.index==gene].any()]]
| tf_occurance | atri_degree_centrality | atri_out_centrality | atri_in_centrality | atri_betweenness_centrality | atri_closeness_centrality | atri_eigenvector_centrality | lrc_degree_centrality | lrc_out_centrality | lrc_in_centrality | lrc_betweenness_centrality | lrc_closeness_centrality | lrc_eigenvector_centrality | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GL2 | 2.0 | 0.115607 | 0.017341 | 0.098266 | 0.005529 | 0.00057 | 0.040031 | 0.111111 | 0.074786 | 0.036325 | 0.52678 | 0.001042 | 0.049802 |
tf_spec.to_csv('TF_GRN_centrality_t7-t9_zscore3.csv', index=True)